A complete mathematical model for logarithmic spiral type sprag one-way clutch design and analysis is given.It assumes that the motion of all clutch components can be expressed by a model of epicyclic gearing.It takes...A complete mathematical model for logarithmic spiral type sprag one-way clutch design and analysis is given.It assumes that the motion of all clutch components can be expressed by a model of epicyclic gearing.It takes advantage of Hunt-Crossley contact impact theory to calculate the contact forces between sprags and races,and it can be used for optimization of design and comparison with other types of sprag clutches.A good deal of analysis shows that the parameters of the steady windup angle,the steady contact force,the natural frequency and natural cycle of clutch have nothing to do with the initial velocity of outer race,while the parameters of the maximum transient windup angle,the maximum transient impact force and the steady engagement time increase linearly in the mode of engaging operation of clutch.It is also shown that the strut angle has great influence on the dynamic engagement performance of clutch.The parameters of the steady windup angle,the maximum transient windup angle,the steady engaging time,the steady contact force,the maximum transient impact force and the natural cycle of clutch decrease linearly nearly with the inner strut angle,while the natural frequency of the system increases linearly with the inner strut angle.展开更多
当前高速铁路接触网参数检测中,存在开口销体积小、分布分散、故障缺陷识别困难,过度依赖综合检测车等问题。本文提出一种采用无人机航拍,结合图像分割与识别技术的基于更快的区域卷积神经网络(Faster R-CNN)算法实现图像处理和优化,进...当前高速铁路接触网参数检测中,存在开口销体积小、分布分散、故障缺陷识别困难,过度依赖综合检测车等问题。本文提出一种采用无人机航拍,结合图像分割与识别技术的基于更快的区域卷积神经网络(Faster R-CNN)算法实现图像处理和优化,进而对开口销缺陷进行检测识别的方法,有效地提升开口销缺陷识别准确率和有效性。测试结果表明,采用基于Faster R-CNN算法的无人机高速铁路接触网开口销缺陷检测方法的开口销图像缺陷识别准确率可达到98%以上,平均精度约90%,接受者操作特征曲线下的面积(area under curve,AUC)大于0.98。该算法通过软件开发工具包(software development kit,SDK)嵌入到无人机,实现接触网开口销自动巡检、智能识别,为现场作业提供智能化检测设备,提升接触网的智能化检测手段,保障高速铁路安全运行。展开更多
基金Project(2011CB706800)supported by the National Basic Research Program of China
文摘A complete mathematical model for logarithmic spiral type sprag one-way clutch design and analysis is given.It assumes that the motion of all clutch components can be expressed by a model of epicyclic gearing.It takes advantage of Hunt-Crossley contact impact theory to calculate the contact forces between sprags and races,and it can be used for optimization of design and comparison with other types of sprag clutches.A good deal of analysis shows that the parameters of the steady windup angle,the steady contact force,the natural frequency and natural cycle of clutch have nothing to do with the initial velocity of outer race,while the parameters of the maximum transient windup angle,the maximum transient impact force and the steady engagement time increase linearly in the mode of engaging operation of clutch.It is also shown that the strut angle has great influence on the dynamic engagement performance of clutch.The parameters of the steady windup angle,the maximum transient windup angle,the steady engaging time,the steady contact force,the maximum transient impact force and the natural cycle of clutch decrease linearly nearly with the inner strut angle,while the natural frequency of the system increases linearly with the inner strut angle.
文摘当前高速铁路接触网参数检测中,存在开口销体积小、分布分散、故障缺陷识别困难,过度依赖综合检测车等问题。本文提出一种采用无人机航拍,结合图像分割与识别技术的基于更快的区域卷积神经网络(Faster R-CNN)算法实现图像处理和优化,进而对开口销缺陷进行检测识别的方法,有效地提升开口销缺陷识别准确率和有效性。测试结果表明,采用基于Faster R-CNN算法的无人机高速铁路接触网开口销缺陷检测方法的开口销图像缺陷识别准确率可达到98%以上,平均精度约90%,接受者操作特征曲线下的面积(area under curve,AUC)大于0.98。该算法通过软件开发工具包(software development kit,SDK)嵌入到无人机,实现接触网开口销自动巡检、智能识别,为现场作业提供智能化检测设备,提升接触网的智能化检测手段,保障高速铁路安全运行。
